VST proper technique question

[Teddy Hallaron]

Working on a film. Will have about 40 or so tracks of dialogue going which I want to apply some BBE VST processing. Should I mix this down to a stereo dialogue stem and then apply once as an over all effect? Set up as an aux? or will I gum up the works by applying it individually to over 40 channels? It would be nice to tweak it a little on a scene by scene basis (instead of the overall approach on the mixdown stem). Your thoughts? btw I've had to do a lot of iZotope noise processing to the original tracks, but that chore is handled discretely within iZotope and new processed files brought back in.

[Dave Labrecque]

Working on a film. Will have about 40 or so tracks of dialogue going which I want to apply some BBE VST processing. Should I mix this down to a stereo dialogue stem and then apply once as an over all effect? Set up as an aux? or will I gum up the works by applying it individually to over 40 channels? It would be nice to tweak it a little on a scene by scene basis (instead of the overall approach on the mixdown stem). Your thoughts? btw I've had to do a lot of iZotope noise processing to the original tracks, but that chore is handled discretely within iZotope and new processed files brought back in.

I think it would depend on the type of processing you're talking about. I guess I've only typically applied compression at the bus level on something like background vocal harmonies, where there are several parts/tracks that are nearly identical. Even then, though, I wonder if I'm sacrificing something.

It would really depend on a lot of factors specific to the content of the tracks, I'd think. How different are they, content-wise? Print-gain wise? Spectrally? Dynamically?

I don't think I've done much sub-mix processing of dialogue tracks when doing film work, FWIW. Too much variation with the source tracks. But, then, I typically get pretty l lousy source tracks to work with. :o

[Hava_wj]

First see if the BBE .....xer VST can be controlled with "automation" through SS. If it can, then the easiest approach would be to use the BBE on the output bus FX. Then each of your 40 MT input tracks can be adjusted for optimal results automatically.

[Teddy Hallaron]

interesting Hava_wj, I'll check the automation compatibility. Also, since each track usually corresponds to a single scene or similar ambient environment later in the film, each track usually represents only a few minutes of screen time; and not all tracks will need BBE processing. So, I'm hoping I can get away with being able to fine tune each track independently if it won't choke the engine. I realize most of this may be RAM and CPU dependent (right?).

And Dave, this one is a doozy for location noise, , , really giving the iZotope a workout.

I'll keep you guys posted on what seems to work the best. Thanks.

[Dave Labrecque]

Unless I'm misunderstanding...

FYI, there's no VST automation in SAW. You can automate the SAW wrapper's bypass control, but that's it. That's often not recommended, however, especially with plugs that have latency.

[Carl G.]

Unless I'm misunderstanding...

FYI, there's no VST automation in SAW. You can automate the SAW wrapper's bypass control, but that's it. That's often not recommended, however, especially with plugs that have latency.

No... but he could 'stack' a few in series and control several different adjustments with enable/disable at appropriate points.

[Dave Labrecque]

No... but he could 'stack' a few in series and control several different adjustments with enable/disable at appropriate points.

So long as they're not latency type plugins, that should work great. Otherwise, not so much, in my experience.

[Teddy Hallaron]

Carl, it looks like the engage/disengage will work in automation. @Dave, don't know if there are latency issues with this technique (I ain't that smart). Actually - not sure if I will need to automate; it seems that with the limited screen time that will be demanded of the VST I may be able to just utilize on a per track basis as needed. I'm finding I do need to tweak a little per scene and so this is hopefully the most sane way to approach it. The adventure continues. . .

[Dave Labrecque]

Carl, it looks like the engage/disengage will work in automation. @Dave, don't know if there are latency issues with this technique (I ain't that smart). Actually - not sure if I will need to automate; it seems that with the limited screen time that will be demanded of the VST I may be able to just utilize on a per track basis as needed. I'm finding I do need to tweak a little per scene and so this is hopefully the most sane way to approach it. The adventure continues. . .

Teddy, you can tell if a plugin induces latency by clicking on the plug's blue triangle and looking at the menu item near the bottom; if the value is zero, then you should be good. If it shows a value other than zero, you've got a latency-inducing plugin. You can always just try automating the bypass and see if it causes trouble, of course.

I don't find that I need to automate VST's all too much, but it would be pretty handy now and again. I do it with the native plug-ins occasionally.

FWIW -- you mention tweaking a little per scene. A few years ago I started breaking film projects up into separate scenes (one EDL per scene). This was a boon in a few ways. First, when I made a major offset goof to an entire track, it wasn't a 90-minute catastrophe (:eek:). Also, mixing each scene out to stems, then re-assembling all scene stems in a master session EDL, allowed great ease and control of transitioning scene to scene. Not sure if you're already doing that, but it would also relieve you quite a bit of the need to automate individual plugins, I'd think. It also reduces the chances of the tragedy of something bad happening to that one EDL that contains EVERYTHING.

[Carl G.]

So long as they're not latency type plugins, that should work great. Otherwise, not so much, in my experience.

Good point if one is building a mix all within the time line of a 'disabled', followed by 'enable' automation routine on a latency pluggin.

When it comes to latency inducing pluggins --- the operator certainly needs to be "Up To Speed" on the best practices! :) :)
(or he is simply "Left Behind the Times" of what is happening "Now")

... I got a million of em' :)
Did you hear about the one.......

[Dave Labrecque]

Good point if one is building a mix all within the time line of a 'disabled', followed by 'enable' automation routine on a latency pluggin.

When it comes to latency inducing pluggins --- the operator certainly needs to be "Up To Speed" on the best practices! :) :)
(or he is simply "Left Behind the Times" of what is happening "Now")

... I got a million of em' :)
Did you hear about the one.......

Probably. :p ;)